The preferential antagonism of pentylenetetrazole proconflict responses differentiates a class of anxiolytic benzodiazepines with potential antipanic action.

With the use of the shock-induced suppression of water drinking in thirsty rats (Vogel's conflict paradigm) and the pentylenetetrazole-enhanced shock-induced suppression of drinking (proconflict paradigm) as animal models to test anxiolytic and antipanic agents, it was possible to distinguish two major classes of benzodiazepines (BZDs) and congeners on the basis of their antiproconflict index (ratio of anticonflict/antiproconflict potencies). Thus, typical low potency BZDs and congeners (diazepam, midazolam, zolpidem, alpidem) with anxiolytic/hypnotic properties have a low antiproconflict index (close to 1), whereas typical high potency BZDs (clonazepam, alprazolam, bretazenil) with reported antipanic properties have an antiproconflict index approximately 10-fold higher. The anticonflict and antiproconflict actions of BZDs with low or high antiproconflict indices are blocked by flumazenil but are potentiated differentially by the gamma-aminobutyric acid (GABA) reuptake blocker 1-2-[bis(trifluoromethyl)-phenyl]-methoxyethyl-1,2,5,6-tetrahydro-3- pyridine-carboxylic acid. Protracted administration of antianxiety and antipanic antidepressant drugs that do not act on GABAA receptors, directly, resulted in anticonflict and antiproconflict effects. However, the efficacy of these drugs is clearly inferior (20-30%) to that of BZDs. These data suggest that specific GABAA receptor subtypes mediate the pharmacological action of BZDs possessing low and high antiproconflict indices